Integrating Sustainable Lighting into Urban Green Space Management: A Case Study of Light Pollution in Polish Urban Parks

Abstract

Urban parks often represent the last viable habitats for wildlife in city centres, functioning as crucial refuges and biodiversity hotspots for a wide array of plant and animal species. This study investigates the issue of light pollution in urban parks in selected Polish cities from the perspective of sustainable urban development and dark-sky friendly ordinances. Field data conducted in 2024 and 2025 include measurements of Upward Light Output Ratio (ULOR), illuminance, luminance, correlated colour temperature (CCT), and spectral characteristics of light sources. In addition, an analysis of changes in the level of light pollution in the studied parks and their surroundings between 2012 and 2025 was performed using data from the VIIRS (Visible Infrared Imaging Radiometer Suite) located on the Suomi NPP satellite. Results highlight the mismatch between sustainable development objectives and the current practice of lighting in most of the analysed parks. The study emphasises the need for better integration of light pollution mitigation in urban spatial policies and provides recommendations for environmentally and socially responsible lighting design in urban parks.

1. Introduction

Urban parks often serve as the only remaining habitats in city centres where wildlife can persist, acting as important refuges and biodiversity hotspots for numerous plant and animal species [1,2,3]. However, urban green spaces are frequently designed with a primary focus on human needs (improving the quality of life for city dwellers, offering spaces for recreation, and relaxation) [4,5] which can have significant negative consequences for the wild organisms inhabiting these environments. One widespread factor aimed at enhancing human safety and comfort in urban parks—artificial light at night (ALAN)—can adversely affect a wide range of light-sensitive species, including plants and animals [6]. It is important to preserve biodiversity, including the health of plants, including tree species growing in parks, because they also support rainwater retention, improve air quality, and influence the local microclimate, thereby providing a range of ecosystem services [7]. Their presence also contributes to cities’ adaptation to climate change and helps reduce the urban heat island effect [8].

Contemporary urban challenges, including urbanisation and the intensification of lighting infrastructure, have led to increased light pollution [9]. This phenomenon refers to a range of adverse effects associated with excessive, unnecessary, or poorly directed light at night, which disrupts natural biological cycles and negatively impacts both environmental quality and human health [9,10,11]. It manifests in several forms, including skyglow, glare, over-illumination, and light trespass—each of which can differently affect urban ecosystems and residents’ well-being. Furthermore, light pollution disrupts the phenological patterns of the growing season more than higher temperatures, especially in urban areas [12]. These issues are often overlooked in spatial planning processes in green areas. However, integrating “dark sky friendly” principles into the design of urban park lighting can significantly mitigate these adverse effects [11]. Therefore, sustainable lighting management in parks requires an interdisciplinary approach that balances the needs of safety, aesthetics, and ecology [12,13,14].

This article focuses on analysing the potential impact of park lighting on the biodiversity and ecosystems in selected Polish city parks, within the broader context of sustainable green space management. The main objectives of the study are to: (1) assess changes in radiance levels in the analysed parks between 2012 and 2025; and (2) identify existing solutions related to sustainable lighting techniques. The spatial scope of the first objective includes 15 parks located in large Polish cities (Kraków, Lublin, Łódz, Toruń, Warsaw, and Wrocław). For the second objective, field studies have been conducted in two selected parks in Kraków and two in Lublin. An additional objective was to analyse the lighting in terms of potential discomfort to park visitors (e.g., excessively bright light).

2. Literature Review

The idea of open city parks is not new. As early as the 15th and 16th centuries, public parks were created for the residents of London, and a little later also in Paris [15]. The establishment of city parks was intended to create a meeting place for residents, but also to encourage them to spend time outdoors, which was supposed to improve their health [16,17]. This was important because the number of city dwellers has been constantly growing for years all over the world [18]. To ensure that they can live comfortably, safely, and with dignity, it is recommended to create green spaces, including large urban parks [19,20,21], but also mini parks, known as pocket parks [22].

City parks currently serve many functions (services) both for city residents and for the environment [23,24,25,26]. In many places, lighting is installed to ensure safe movement around the park. However, this lighting should be adjusted to the specific location. In terms of intensity, it cannot be stated unequivocally that a more brightly lit path in a park will necessarily be safer or more user-friendly, because excessive light in one place can limit perception of the surroundings and reduce the sense of safety [27,28]. To counteract such problems, some of the lights in parks are switched off when not in use and switched on when needed (motion sensors can be installed). Another solution is to dim the lights late at night.

Also, the colour temperature of lights in parks can have adverse effects [29]. For example, different colour temperatures are preferred by visitors at restrooms or on pathways in parks (3700 and 4000 K, respectively) but cool white light (6000 K) is undesirable. In addition to taking into account the technical issues mentioned above, it is important to also introduce unlit (dark) areas, which are places not intended for people but for other organisms (animals) [1]. In spatial planning, in addition to green and blue infrastructure, there are increasing mentions of dark infrastructure—as an unlit or very dimly lit space that ensures the functioning of many animal species, also in urban parks [30].

Sustainable lighting seeks to minimise the adverse effects of ALAN by designing and implementing lighting solutions that meet users’ needs (such as safety, spatial orientation, and aesthetics) while taking into account potential environmental consequences [10,11,12]. One of the key technical parameters influencing light pollution is the Upward Light Output Ratio (ULOR)—an indicator of the percentage of light emitted into the upper hemisphere. As ULOR increases, skyglow intensifies, disrupting the movement of wild animals at multiple scales: it increases the risk of collisions with illuminated structures, alters stopover site selection and the aerial connectivity of the night sky, and ultimately affects phenology [31]. The lower the ULOR, the smaller the contribution to skyglow and energy waste [32]. Another important parameter is Correlated Colour Temperature (CCT)—the colour temperature of light, expressed in Kelvins. Research shows that light sources with high CCT (e.g., >4000 K, typical of white or blue light) have stronger biological and ecological impacts than warmer light sources (<3000 K) [33], as the blue spectrum more significantly disrupts the circadian rhythms of animals [10,11]. In addition, blue light is scattered by the atmosphere more strongly than other components of white light, which results in a greater increase in the brightness of the night sky. Therefore, light sources with higher CCT cause a stronger skyglow effect [34].

Taking the above-mentioned key guidelines into account in park lighting is crucial, because numerous studies have shown disturbances in the behaviour of nocturnal animals (e.g., insects, bats, birds) caused by ALAN, as well as abnormal development and phenology of plants, including in urban parks [13,35,36,37,38]. Therefore, it is important that light should illuminate paths and enable safe movement, but it should not illuminate plants or animal habitats. Light directed at the crowns of trees or shrubs not only impairs their functioning and reduces their lifespan [37,39,40,41], but also affects, for example, birds nesting in tree crowns, small mammals like hedgehogs or insects living there [42,43,44,45,46]. Upward-facing lighting should be avoided, and light sources such as ‘glowing balls’ should be replaced with proper ones.

Regarding the importance of CCT in urban green areas, studies examining the wavelength sensitivities of various terrestrial animals have shown that longer wavelengths offer the best balance between supporting human visual performance at night and minimising visual disturbance to other species. Most animals and plants have various photoreceptors that allow them to gather information about their surroundings and make them sensitive to different wavelengths of light [47,48]. Therefore, reducing CCT is especially beneficial for many wildlife species and plants by helping to mitigate the negative effects of light pollution in combination with dimming, shielding, or part-night lighting [49]. In order to limit the negative impact of park lighting on living organisms, low radiation intensities should be used, preferably light sources with a warm colour temperature—low in blue radiation. It is recommended to use sources with a temperature of 2700–3000 K or radiation known as amber in parks.

Eliminating artificial lighting from urban green areas may prove difficult, however, as social perceptions of light are complex [50]. Lighting can enhance users’ sense of safety and comfort after dark. Nevertheless, its overuse or the application of excessively “cool” (bluish) light may detract from the aesthetic quality of the space, create visual dissonance, and ultimately diminish the potential of parks as places of recreation, relaxation, and contact with nature [51,52]. For these reasons, the implementation of sustainable lighting—alongside other environmental considerations—is gaining significance in comprehensive urban planning [53,54].

In recent years, tools such as lighting master plans have emerged across Europe and North America, integrating aesthetic, technical, and ecological aspects into urban lighting design [6,55,56]. Local governments often use guidelines from international organisations such as Commission Internationale de l’Éclairage (CIE) and DarkSky International. According to the guidelines of both organisations, minimising ULOR to 0% is optimal in most cases for reducing skyglow. However, in some applications (e.g., illumination of objects), upward-directed light is acceptable. DarkSky International recommends the use of light sources with CCT < 3000 K (so-called ‘warm white’) or LEDs with a filter limiting short-wave emission to minimise the negative impact of blue light on ecosystems, human health, and skyglow levels. CIE does not specify a direct CCT limit, but emphasises the role of spectrum selection in limiting Rayleigh scattering [32,57].

Taking into account the above literature review, our study will fill a research gap, enabling us to propose recommendations strictly adapted to the characteristics of urban parks (Figure 1).

Examples from cities such as Amsterdam, Ottawa, Toronto, and Vienna demonstrate that it is possible to combine modern LED technologies with dark sky protection principles by, for instance, limiting ULOR, reducing CCT, automatically dimming lights at night, or establishing controlled lighting zones. In Vienna, a special LED luminaire was developed to unify the light sources in the city. It has ULOR 0% and CCT 4000 K. Since 2017, over 80,000 luminaires have been replaced, and the entire investment is to be completed in 2026. The first energy-efficient luminaires in this city were installed in city parks [58]. Another example is Ottawa, where different lighting zones were designated and different fixtures were used to illuminate buildings in the 2200–3200 and 3500–4000 K ranges. In addition, between 2019 and 2023, more than 23,000 streetlamps were replaced with LEDs, with additional capacity for a smart grid and controls. The remaining 10,000 are to be replaced in the coming years. The investment has already resulted in reduced annual electricity consumption by approximately 13 M kWh [59].

3. Materials and Methods

3.1. Research Area

Poland is one of the countries where the implementation of light pollution mitigation policies is particularly urgent. First, Polish cities have been intensifying efforts to modernise their lighting infrastructure for over a decade, primarily by replacing sodium–vapour luminaires with LEDs—a shift that brings energy savings, but also new challenges related to increased blue-spectrum light pollution. Second, these issues have not yet been widely addressed in national regulations, and the integration of sustainable lighting principles into park design remains rather uncommon [60]. Furthermore, satellite data indicate that Poland is among the countries experiencing a notable and increasing trend in light pollution levels, particularly in urban and suburban areas [9,61]. The context of Polish urban parks—often located within densely built-up environments yet performing essential ecological functions—thus offers a valuable case for analysing the potential for integrating sustainable lighting into urban greenery management. Especially given the lack of nationwide regulations minimising the negative impact of light pollution.

During the study 15 parks, located in large Polish cities (Kraków, Lublin, Łódź, Toruń, Warsaw, and Wrocław) were chosen (

Table 1. Characteristics of the analysed urban parks in Poland.

No	City (and Its Population in Thousands)	Name of the Park (in English)	Name of the Park (in Polish)	Area [ha]	Year of Creation	Measurements
						Satellite	Field
1	Kraków (809.2)	Kraków Park	Park Krakowski	5.34	1885	Yes	Yes
2		Jerzmanowski Park	Park Jerzmanowskich	7.00	1810	Yes	Yes
3		Polish Airmen Park	Park Lotników Polskich	43.06	1966	Yes	No
4		Jordan Park	Park Jordana	19.77	1889	Yes	Yes
5	Lublin (328.3)	Saski Garden	Ogród Saski	12.80	1837	Yes	No
6		People’s Park	Park Ludowy	23.00	1951	Yes	Yes
7	Łódź (645.7)	Poniatowski Park	Park Poniatowskiego	41.60	1910	Yes	No
8		Źródliska I Park	Park Źródliska I	10.64	1840	Yes	No
9	Toruń (193.7)	City Park	Park Miejski	25.00	1817	Yes	No
10	Warsaw (1863.8)	Skaryszewski Park	Park Skaryszewski	58.00	1905	Yes	No
11		Mokotowskie Field	Pole Mokotowskie	73.00	1916	Yes	No
12		Royal Baths	Łazienki Królewskie	76.00	1767	Yes	No
13		Szczęśliwicki Park	Park Szczęśliwicki	30.11	1961	Yes	No
14	Wrocław (672.9)	Popowicki Park	Park Popowicki	16.50	1897	Yes	No
15		Szczytnicki Park	Park Szczytnicki	112.00	1783	Yes	No

, Figure 2). They included (in terms of radiance measurements) the following parks (names in Polish): Krakowski, Jerzmanowskich, Lotników Polskich, and Jordana in Kraków; Ogród Saski and Ludowy in Lublin; Poniatowskiego and Źródliska I in Łódź; Miejski in Toruń; Skaryszewski, Pole Mokotowskie, Łazienki Królewskie, Szczęśliwicki in Warsaw; as well as Popowicki and Szczytnicki in Wrocław. The English and Polish names of the parks surveyed are listed in Table 1. In the article, we will use the Polish names.

Satellite imagery surveys were conducted for 15 sites in Poland. This allowed us to develop a research methodology that can be used for all other parks in Poland and abroad. Satellite data is available for any selected area, so we decided to present selected parks to show their variability over the years. We chose large parks located in the vicinity of the centres of large cities.

Field research was carried out in four selected locations: in Krakowski and Jerzmanowskich parks in Kraków, and in Ogród Saski and Ludowy parks in Lublin, demonstrating the possibility of use and outlining a vision for future development. When selecting these four parks, we also took into account their importance as local ecosystems, including their role as habitats for birds nesting in tree crowns (between 34 and 61 species of birds live in the parks surveyed) (Appendix A). In addition, they are inhabited by species of bats, small mammals and reptiles, as well as many species of insects. They are also home to many varieties of trees, shrubs, and perennials.

3.2. Radiance Data Analysis

The research process consisted of two main stages. The first stage involved analysing radiance levels and their changes between 2012 and 2025 in the areas of the studied parks. The research material consisted of data derived from Visible Infrared Imaging Radiometer Suite (VIIRS) observations. VIIRS is a radiometer installed on board the Suomi NPP polar-orbiting satellite, with equatorial crossing times at 01:25 CET (night) and 13:25 CET (day). Specifically, we used the standard product VNP46A2, which contains radiance data corrected for lunar illumination effects and adjusted for the presence of snow and cloud cover (“VIIRS/NPP Gap-Filled Lunar BRDF-Adjusted Nighttime Lights Daily L3 Global 500 m Linear Lat/Lon Grid”) [62].

Radiance values (vertical axis, nW·cm⁻²·sr⁻¹) are presented in Figure 3, Figure 4, Figure 5 and Figure 6 in Section 4.1. and Figure A1, Figure A2, Figure A3, Figure A4, Figure A5, Figure A6, Figure A7, Figure A8, Figure A9, Figure A10 and Figure A11 in Appendix B in two forms. First, individual cloud-free nights between 2012 and 2025 are shown as black dots, representing instantaneous radiance measurements for a given VIIRS pixel. Second, all instantaneous observations within a given year are aggregated to compute the annual median radiance, shown as a red line spanning the corresponding year, and annual mean, shown as a blue line. For certain dates, radiance values deviate substantially from the typical range, which justifies the use of the median—a statistic less sensitive to outliers—over the mean.

The coarse spatial resolution of the VIIRS instrument introduces the possibility that a single pixel may extend beyond the boundaries of a park—particularly when the park’s dimensions are smaller than the VIIRS footprint (approximately 460 m × 300 m in the study area). For this reason, the data may locally represent the impact of both the park’s lights and the surrounding illumination. However, such lighting outside the park may affect the vegetation inside the park and emphasises that the parks should adhere to good lighting practices in their immediate surroundings. Other limitations of VIIRS data are insensitivity to blue light (e.g., from LED sources) and the potential impact of tree canopy (blocking the light). As no alternative dataset provides both homogeneous radiance information and a multi-year time series, we opted to include VIIRS data despite its limitations.

3.3. Field Measurements

The second stage of the research process comprised field research conducted in 2024 and 2025. This included measurements of illuminance (measured with a high-sensitivity luxmeter), luminance (captured using a matrix luminance camera), correlated colour temperature (CCT), spectral characteristics of light sources (measured with a spectrometer), and Upward Light Output Ratio (ULOR) using a visual assessment method. Since estimating ULOR by eye can be difficult and fraught with considerable uncertainty, we limited ourselves to evaluating light sources using the popular cutoff classification: cutoff, semi-cutoff (typical park lamp) and non-cutoff (ball/globe lamp). The limitation of this method is that it does not take into account incorrect installation of a lamp. A cutoff and semi-cutoff luminaire installed at an angle not parallel to the ground will have the ULOR above 0%

A Sonopan L-200 precision lux metre, meeting the requirements of class A according to CIE, with the spectral correction of the photometric head corresponding to the spectral sensitivity function for photopic vision V(λ), was used to measure illuminance. The importance of the metrological quality of photometers in the context of the reliability of measurement results is presented, among others, in the works [63,64]. Luminance measurements were performed with a GL Opticam 3.0 4K TEC matrix metre, consisting of a high-resolution camera (4096 × 2160 pixels) with a sensor thermal stabilisation system and a set of interchangeable lenses. The spectral sensitivity of the device is adjusted to the V(λ) function. The spectral characteristics of the light emitted by the luminaires were recorded using a GL SPECTIS 5.0 Touch spectrometer. The technical parameters of the devices are presented in

Table 2. Technical parameters of measuring devices used in field measurements. Source: own elaboration.

Device	Manufacturer	Measured Quantity	Measuring Range	Resolution	Measurement Error
Illuminance metre L200	Sonopan (Białystok, Poland)	Illuminance	0.001–50 lx	0.001 lx	±3% indications ±1 digit (2856 K)
			0.1–5000 lx	0.1 lx
			10–500,000 lx	10 lx
Imaging luminance metre GL Opticam 3.0 4 K TEC	GL Optic (Puszczykowo, Poland)	luminance	0.005–5,000,000 cd/m2	4 K (4096 × 2160 px)	±5%
Spectroradiometer GL SPECTIS 5.0 Touch	GL Optic (Puszczykowo, Poland)	spectral characteristics	340–830 nm	0.5 nm (spectral)	±0.5 nm

. Photos of selected illuminated objects in parks were also taken and attached to this work to illustrate cases of incorrect use of lighting.

The additional stage involved an analysis of local strategic documents and environmental protection programmes in the cities of Kraków, Lublin, Łódź, Toruń, Warsaw, and Wrocław. This analysis aimed to identify references to light pollution in the context of urban greenery management. Additionally, it sought to correlate changes in radiance observed in the studied parks with the implementation timelines of sustainable lighting practices described in these documents.

4. Results

4.1. Changes in Radiance

The analysis of radiance changes in the four parks in Kraków and Lublin where field studies were later conducted (Figure 3, Figure 4, Figure 5 and Figure 6), as well as in the remaining parks (Appendix B), shows that in most cases there were no significant changes in radiance levels in the examined period of time. Exceptions include Park Jordana in Kraków, where the lighting is turned off after park closure (from 10 p.m.), and Park Krakowski P, also in Kraków, which underwent a lighting upgrade to LEDs in 2017. In both cases, a decrease in average radiance was observed during the analysed period. However, similar decreases were not observed in other parks that had been revitalised and equipped with new lighting systems—for example, Park Ludowy in Lublin, modernised in 2020. One reason is the installation of a much larger number of lighting fixtures compared to the situation before the revitalisation. The modernisation often covered not only the park area but also the surrounding streets, which, due to the imperfections of VIIRS measurements (pixel size), affected the results obtained. One such case may be Park Szczęśliwicki in Warsaw.

Figure 3. Changes in radiance in 2012–2024 in Park Krakowski, Kraków. Source: own elaboration.

Figure 3. Changes in radiance in 2012–2024 in Park Krakowski, Kraków. Source: own elaboration.

Figure 4. Changes in radiance in 2012–2024 in Park Jerzmanowskich, Kraków. Source: own elaboration.

Figure 4. Changes in radiance in 2012–2024 in Park Jerzmanowskich, Kraków. Source: own elaboration.

Figure 5. Changes in radiance in 2012–2024 in Ogród Saski, Lublin. Source: own elaboration.

Figure 5. Changes in radiance in 2012–2024 in Ogród Saski, Lublin. Source: own elaboration.

Figure 6. Changes in radiance in 2012–2024 in Park Ludowy, Lublin. Source: own elaboration.

Figure 6. Changes in radiance in 2012–2024 in Park Ludowy, Lublin. Source: own elaboration.

The occasional declines in radiance observed in the studied parks cannot be attributed to a deliberate policy aimed at reducing light pollution in urban green areas. This is supported by the lack of references to light pollution in local development strategies and environmental protection programmes in Kraków and Lublin. Nevertheless, the concept of sustainable development is present in these documents. When it comes to lighting, energy efficiency is frequently mentioned and often cited as the primary reason for replacing sodium lamps with LEDs. In other cities, strategic documents also do not contain any references to LP and ALAN. However, in two cities (Warsaw and Wrocław), there are references to this phenomenon in local environmental protection programmes. In Wrocław, the information is limited to laconic mentions that LP has a negative impact on animals, plants, and people [65], but in Warsaw, the entire chapter on LP contains guidelines and recommendations about ULOR, CCT, and other parameters concerning lighting not only on the streets, but also in urban parks and squares [66].

4.2. Field Surveys

Aside from obvious cases of high upward light emission, such as “glass globe” luminaires (all lamps in Ogród Saski in Lublin), which have been shown to exhibit a significant proportion of luminous flux emitted into the upper hemisphere (approx. 60% goes up) [67], most parks were equipped with LED fixtures featuring semi-cutoff lamps, and with flat glass (cutoff) or other design elements that helped reduce skyward light loss. However, the measured illuminance on park paths varied significantly. For instance, in Park Krakowski (Kraków), it ranged from 1.8 lx at the centre of the path to over 37 lx directly beneath the lamps (Figure 7). In Park Jerzmanowskich (Kraków), a similar section yielded readings from 2 lx to just over 21 lx (Figure 8 and Figure 9). Comparable variations were also observed in luminance measurements (Figure 10 and Figure 11).

The requirements for lighting parameters in park paths are defined in the European standard EN 13201-2:2015 (Road Lighting—Part 2: Performance Requirements), which provides lighting classes, “P”, for roads and traffic routes with low traffic intensity, including pedestrian and bicycle paths [68]. This standard distinguishes six P-type lighting classes: P1, P2, P3, P4, P5, and P6. The individual classes differ in terms of minimum and average lighting intensity requirements. The higher the lighting class, the higher the lighting requirements. For example, in the case of lighting class P1, the average illuminance value on a park alley (E_mean) should not be less than 15 lx, while the minimum value (E_min) should be at least 3 lx.

An analysis of the measurement results obtained in three parks allowed them to be assigned to the appropriate lighting classes. In the case of the park whose data is shown in Figure 7, the minimum lighting intensity value E_min = 4.65 lx and the average lighting intensity value E_mean = 14.5 lx. This corresponds to the requirements of class P2, which requires E_mean ≥ 10 lx, E_min ≥ 2 lx. For the next park (Figure 8), the obtained values E_mean = 7.84 lx and E_min = 3.05 lx meet the requirements of class P3 (≥7.5 lx and ≥1.5 lx). In the case of the third park (Figure 9), the obtained values of E_mean = 9.66 lx and E_min = 1.79 lx also meet the requirements of class P3.

The most pronounced differences were found in the spectral characteristics and correlated colour temperature (CCT) of the lighting. In some parks, such as the Ogród Saski in Lublin, “glass globe” luminaires featured low CCT and minimal blue light emission (Figure 12). In contrast, Park Jerzmanowskich used fixtures with CCT values above 3000 K and a notable blue light component (Figure 13). Park Krakowski exhibited luminaires with CCT values exceeding 4000 K and a dominant blue light share (Figure 14). These figures show examples of spectral characteristics for the selected, most common types of lighting fixtures for those parks. Their purpose was to show the differences in spectral characteristics between individual luminaires, including the proportion of blue components in light with a higher colour temperature.

In addition to the measured lighting parameters, several apparent contradictions with the principles of sustainable lighting were observed in the studied parks. Many such inconsistencies were found in Park Ludowy in Lublin. One example was glare, particularly noticeable along illuminated staircases on the main avenue, as well as on an excessively lit pedestrian bridge (Figure 15). In addition to causing discomfort for users, such installations attract insects, contributing to declines in their populations (Figure 16). Moreover, many luminaires were installed high in the tree canopy, resulting in energy waste (since the ground below remained dark) and directing beams of light directly onto tree branches, which may negatively affect the plants (Figure 16 and Figure 17).

Therefore, it is very important to monitor the impact of lighting, taking into account detailed measurements that go beyond standard light intensity measurement because the human visual system does not directly perceive light intensity. Light intensity is a photometric quantity commonly used in lighting technology to characterise the amount of luminous flux falling on a given surface. It is an objective parameter, expressed in lux, which can be relatively easily measured using widely available and inexpensive instruments. What humans actually perceive—and what is closer to their subjective feelings—is luminance. For example, two surfaces may have identical illuminance values, but due to differences in reflectance (e.g., light-coloured pavement and dark asphalt), their luminance will be different. Luminance, expressed in cd/m², is a measure of the perceived brightness of a surface.

In practice, for areas such as park paths or pavements, the basic design and evaluation criterion is illuminance, mainly due to the simplicity and low cost of measurement. Luminance measurements require specialised and much more expensive equipment, which is why they are not standard in these applications and are not required by regulations. Nevertheless, it is luminance that is directly related to the sensation of visual discomfort or glare. The luminance differences of 1–4 cd/m² observed in the study are insignificant and do not usually cause significant discomfort in pedestrian traffic conditions, unless they are accompanied by disturbing glare resulting, for example, from the use of unshielded or improperly installed lighting fixtures (e.g., shown on Figure 15 and Figure 16).

5. Discussion

5.1. Radiance

Although the increase in radiation in Poland in recent years has been one of the highest in Europe [9,61], the data for the parks analysed do not follow this trend. In some, especially in Park Krakowski, there have been declines, while in others, either radiation levels have remained at a similar level or a slight decline has been recorded in the post-COVID-19 pandemic period, as in Park Jordana in Krakow, Park Szczęśliwicki in Wrocław and Park Miejski in Toruń. The decline in the post-pandemic period may have had various causes, including the switching off of street lighting after midnight to save costs, and was noticeable in some areas of Poland [38].

As a result of the failure to consider LP as an environmental problem in strategic documents and environmental protection programmes in some of the cities surveyed (Kraków, Lublin, Łódź, Toruń), any changes in lighting should not be directly linked to deliberate measures to minimise LP, e.g., by switching off lighting at night or changing lighting technologies to dark sky friendly ones. One of the reasons for the partial decrease in radiance in parks may also be the growth of vegetation, especially trees, which can block some of the light emitted in parks. This can be seen in Figure 16 and Figure 17, where park lanterns are surrounded by tree crowns or located directly underneath them. Another reason may be the revitalisation of lighting in parks and its replacement with LED lighting, which means that VIIRS does not detect any change, or even detect a decrease, when in reality the opposite may be true. Such situations have been observed previously when switching from HPS to LED lighting, which led to an apparent decrease in LP as seen by VIIRS, but ground measurements showed an increase in the urban glow. An example of how a simple 1:1 replacement of old lamps with LED lamps is not sufficient to reduce urban glow is Chelton County in Washington State [69].

The VIIRS’s blindness to blue light is the main limitation of the method using its data, along with pixel size, which in some cases causes radiation from areas adjacent to parks to be included. An alternative could be regular surveys and monitoring conducted from drones, but no one has yet performed this for large areas such as city parks due to the cost and time-consuming nature of such surveys.

Failure to consider LP as a risk by municipal authorities stems from a nationwide disregard for the effects of LP, resulting in ALAN not being included in the Polish Environmental Protection Act as a potential harmful factor [60]. However, this does not mean that there are no local initiatives aimed at regulating ALAN and introducing sustainable public lighting, including lighting in urban parks. The analysis of the documents shows that Wrocław [65] and, in particular, Warsaw [66] try to mitigate negative effects of LP. In the latter, direct solutions are proposed, such as the use of ULOR 0%, or power reducers that reduce light emissions during periods of low pedestrian traffic. The guidelines also include the use of light intensity according to the time of day (after 11 p.m., maximum dimming or switching off of lighting until dawn). These solutions are part of possible efforts to reduce light escape into the upper semi-sphere, as recommended by international organisations [57], which would lead to a decrease in the level of radiance from urban parks, thereby contributing to a reduction in the urban glow. Such reductions have already been achieved in various places. Examples include Tucson, a city in Arizona, where over 18,000 HPS/LPS luminaires were replaced with 3000 K colour temperature LEDs with full shielding, resulting in a reduced total luminous flux—approximately 142 million lumens compared to the previous 445 million lm. Modelling predicted a 10–20% reduction in skyglow, which was confirmed by satellite measurements—upward radiance decreased by 7% [70].

5.2. Sustainable Lighting Techniques vs. Ecosystems in Parks

Field data revealed numerous examples of ineffective lighting design, including luminaires placed in treetops (e.g., Figure 16), excessive glare, and the use of higher CCT than 3000 K (Figure 14)—all of which conflict with the principles of sustainable urban lighting recommended by international organisations [32,57]. It should be emphasised that the lighting fixtures on the lanterns in the four parks studied were of various types. The lowest CCT (2150 K) and the smallest proportion of blue light were found in the fixtures in Ogród Saski in Lublin, but due to the design of these fixtures (“glass globe”), they were the least energy-efficient (approx. 60% of the light is emitted upwards) and shone directly into the tree crowns.

Semi-cutoff luminaires were used in the other parks, but they differed in terms of power, CCT and spectral distribution. In Park Krakowski, the CCT of the luminaires was high at 4018 K, while it was slightly lower in Park Jerzmanowskich (3156 K). Despite the use of semi-cutoff models, a large number of them were located in tree crowns or in their immediate vicinity (Figure 16 and Figure 17), thus constituting a direct source of LP for any birds nesting in trees and for ecosystems in general.

In this context, it is particularly important to highlight ULOR and CCT as key technical parameters for assessing the environmental impact of ALAN on ecosystems and the night sky. Observations confirm that even modern LED luminaires, if poorly installed or characterised by high CCT, can negatively affect the environment [12,32,35]. This issue was also described by Pawson and Bader [26], who demonstrated that retrofitting lighting systems without revising the underlying design strategy can lead to ecologically counterproductive outcomes.

Although energy efficiency is frequently cited in analysed urban planning documents as the main motivation for replacing sodium lamps with LEDs, there is no reflection on the potential increase in blue light emissions—now widely recognised as biologically harmful [10,11]. For this reason, the implementation of new technologies should be accompanied by regulations concerning colour temperature, light direction, and intensity. Examples from other cities show that integrating “dark sky friendly” principles is not only feasible but can also improve the visual quality and safety of urban spaces [6,50]. Cities such as Ottawa and Vienna have adopted lighting master plans that enable the controlled and coordinated management of artificial light at the local level [58,59]. This type of strategy—supplemented by cooperation with organisations like Dark Sky International—could serve as a valuable model for Polish cities.

Warsaw could also be an example for other Polish cities of how to try to mitigate the negative effects of LP. In the municipal environmental protection programme [66] direct solutions are proposed, such as the use of ULOR 0%, CCT at a maximum level of 3000 K, or reducing light power during periods of low pedestrian traffic. The guidelines also include the use of light intensity according to the time of day (after 11 p.m., maximum dimming or switching off of lighting until dawn). The Warsaw criteria also directly address urban ecosystems and indicate the need to use lighting that is neutral to animals, with low or zero UVB spectrum emissions, and not to direct lighting straight at plants (especially tree crowns) or water reservoirs. It is also suggested that night-time lighting of buildings be reduced to the minimum necessary during the spring and autumn bird migration periods (from 1 February to 30 May and from 1 August to 31 October). These regulations are in line with the guidelines of international organisations [57] and solutions implemented in other cities [58,59].

However, the other cities surveyed do not have any guidelines on minimising the effects of LP, including on urban ecosystems, which supports the conclusions of Kyba et al. [11], who noted that ecological aspects of artificial lighting remain underrepresented in urban planning, despite their well-documented impacts on the environment and human health [9,10,36].

Although we do not have any data on visitor numbers in the parks surveyed, split by day and time after dark, it can be assumed that there are significantly fewer people in the park in the evening than during the day. Nevertheless, it is important to manage park lighting in order to achieve the right balance between visitor safety, their comfort in terms of the lighting used (no unpleasant light) [50], and the needs of ecosystems. Unfortunately, our observations showed that the power and colour of the light were not always suited to the needs of pedestrians, which in some cases (e.g., Figure 15 and Figure 16) was excessive and could be perceived as disruptive.

6. Conclusions

Designing urban parks that are functional for both humans and wildlife is a significant contemporary challenge, particularly in the context of climate change and biodiversity loss. Therefore, urban green areas should be planned in a way that ensures human safety while also supporting the needs of other organisms, which in turn provide a range of essential ecosystem services [71]. The results of the study clearly show that no decrease in radiance was observed in most of the parks studied. In some of them, a decrease was observed by VIIRS, but in order to confirm this definitively, ground-based measurements of night sky brightness would need to be carried out.

This is especially important for those parks where the lighting has been replaced with LED. Unfortunately, there are no results of ground measurements from before the revitalisation of the parks, so it is not possible to compare the data. Therefore, we must take into account the possibility of apparent changes resulting from the imperfections of VIIRS [69]. An additional limitation of the usefulness of VIIRS data in relation to urban parks is the partial blocking of light by tree crowns, which may render the results unreliably. The pixel size, which means that some of the light from neighbouring streets is also taken into account in the analysis, is another limitation of this method in the case of smaller parks.

Ground measurements provided information about the diversity of luminaires used in the park. Only a small proportion had a CCT below 3000 K (Ogród Saski in Lublin), and those that did were non-cutoff. The rest had CCTs in the range of approx. 3100–4100 K, and although they were mostly semi-cutoff, they were often placed in tree crowns and did not function properly, directly disrupting ecosystems. It can be concluded that the cause of these irregularities may be the failure to consider LP as something negative in the analysed local strategic documents.

Ready-made recommendations on how to illuminate in a sustainable manner have existed for years, as exemplified by the references provided in the text, e.g., the Dark Sky International guidelines [57] or the guidelines from the Warsaw Environmental Protection Programme [66]. The failure to include these guidelines in local documents in other cities and the lack of implementation of such solutions in practice may therefore be mainly due to a lack of awareness among local authorities about the harmful effects of LP.

Our study can be considered preliminary, and further research may focus on: (1) investigating the reasons for the lack of interest among local authorities in implementing sustainable lighting in urban parks; (2) changes in the biodiversity of urban ecosystems following the revitalisation of park lighting; (3) ground-based research in selected parks on night sky brightness, supplementing data from VIIRS.